AMERICANMETEOROLOGICALSOCIETY

Bulletin of the American Meteorological Society

EARLY ONLINE RELEASEThis is a preliminary PDF of the author-producedmanuscript that has been peer-reviewed and accepted for publication. Since it is being postedso soon after acceptance, it has not yet beencopyedited, formatted, or processed by AMSPublications. This preliminary version of the manuscript may be downloaded, distributed, andcited, but please be aware that there will be visualdifferences and possibly some content differences between this version and the final published version.

The DOI for this manuscript is doi: 10.1175/BAMS-D-13-00238.1

The final published version of this manuscript will replacethe preliminary version at the above DOI once it is available.

© 201 American Meteorological Society 4

1

Cloud Banding and Winds in Intense European Cyclones: Results from the DIAMET Project 2

G. Vaughan1, J. Methven4, D. Anderson11, B. Antonescu2 , L. Baker4, T. P. Baker7, S. P. Ballard9, K. N. Bower2,

P. R. A. Brown10, J. Chagnon3, T. W. Choularton2, J. Chylik8, P. J. Connolly2, P. A. Cook8, R. J. Cotton10 , J. 4

Crosier1, C. Dearden2, J. R. Dorsey1, T. H. A. Frame3, M. W. Gallagher2, M. Goodliff4, S. L. Gray4, B. J. Harvey4,

P. Knippertz12, H. W. Lean9, D. Li9, G. Lloyd2, O. Martínez –Alvarado4, J. Nicol3, J. Norris2, E. Öström10, J. 6

Owen7, D. J. Parker7, R. S. Plant4, I. A. Renfrew8, N. M. Roberts9, P. Rosenberg7, A. C. Rudd4, D. M. Schultz2, J.

P.Taylor10, T. Trzeciak7, R. Tubbs9, A. K. Vance10, P. J. van Leeuwen5, A. Wellpott11, A. Woolley11 8

AFFILIATIONS: 10

1 National Centre for Atmospheric Science (NCAS) , University of Manchester, Manchester, UK

2 Centre for Atmospheric Science, University of Manchester, Manchester, UK 12

3 NCAS, University of Reading, Reading, UK

4 Department of Meteorology, University of Reading, Reading, UK 14

5 National Centre for Earth Observation (NCEO), University of Reading, Reading, UK

6 NCAS, University of Leeds, Leeds, UK 16

7 School of Earth and Environment,University of Leeds, Leeds, UK

8 School of Environmental Science, University of East Anglia, Norwich, UK 18

9 Met Office, University of Reading, Reading, UK

10 Met Office, Exeter, UK 20

11 Facility for Airborne Atmospheric Measurement, Cranfield, UK

12 Institute for Meteorology and Climate Research, Karlsruhe Institute of Technology, Germany 22

1Corresponding address: SEAES, Simon Building, Oxford Rd, Manchester, M13 9PL, UK 24

26

2

Abstract

The DIAMET (DIAbatic influences on Mesoscale structures in ExTratropical storms) project aims to 28

improve forecasts of high-impact weather in extratropical cyclones through field measurements,

high-resolution numerical modeling, and improved design of ensemble forecasting and data 30

assimilation systems. This article introduces DIAMET and presents some of the first results. Four

field campaigns were conducted by the project, one of which, in late 2011, coincided with an 32

exceptionally stormy period marked by an unusually strong, zonal North Atlantic jet stream and a

succession of severe windstorms in northwest Europe. As a result, December 2011 had the highest 34

monthly North Atlantic Oscillation index (2.52) of any December in the last 60 years. Detailed

observations of several of these storms were gathered using the UK’s BAe146 research aircraft 36

and extensive ground-based measurements. As an example of the results obtained during the

campaign, observations are presented of cyclone Friedhelm on 8 December 2011, when surface 38

winds with gusts exceeding 30 m s-1 crossed central Scotland, leading to widespread disruption to

transportation and electricity supply. Friedhelm deepened 44 hPa in 24 hours and developed a 40

pronounced bent-back front wrapping around the storm center. The strongest winds at 850 hPa

and the surface occurred in the southern quadrant of the storm, and detailed measurements 42

showed these to be most intense in clear air between bands of showers. High-resolution ensemble

forecasts from the Met Office showed similar features, with the strongest winds aligned in linear 44

swaths between the bands, suggesting that there is potential for improved skill in forecasts of

damaging winds. 46

Capsule: New aircraft measurements, together with high-resolution modeling, reveal fine-scale 48

wind structure in an intense extratropical windstorm.

50

3

Extratropical cyclones approaching western Europe along the North Atlantic storm track are a

major cause of damaging winds and heavy precipitation. A particular problem in forecasting these 52

cyclones is that the highest-impact weather within them arises from mesoscale structures such as

fronts and bands of strong winds. These structures are influenced by diabatic processes (those 54

which add or remove heat from the air) such as latent heating and cooling associated with phase

changes of water, fluxes of heat and moisture from the Earth’s surface, and radiative flux 56

convergence. Key elements in diabatic processes are turbulence, convection and cloud physics –

small-scale phenomena which cannot be represented explicitly in numerical weather prediction 58

models. They must therefore be parameterised, introducing a source of systematic uncertainty in

the models. Detailed observations of real events are needed to test the models and ultimately to 60

improve the parameterisation of small-scale processes.

Here we report on initial results from the DIAMET (DIAbatic influences on Mesoscale structures in 62

ExTratropical storms) project, which aims to improve our understanding and predictions of

mesoscale structures within extratropical cyclones by means of field measurements, high-64

resolution modeling and improved design of ensemble forecasting and data assimilation systems.

The modeling component of the project includes evaluation of the Met Office high-resolution 66

ensemble forecasts, the probability forecasts of mesoscale structures, the stochastic physics

scheme used in the ensemble, and the ability of the model to represent the structures that bring 68

high-impact weather. This paper provides an overall introduction to DIAMET with particular

emphasis on the field measurements and the way these are used to test and improve models. 70

The winter campaign of DIAMET was conducted during a period of particularly intense storm 72

activity over the North Atlantic. As an example of an intense storm, and to illustrate the scientific

approach used in DIAMET, we present a more detailed study of Cyclone Friedhelm on 8 December 74

4

2011. The strongest low-level winds in this storm occurred on its southern and southwestern

flanks. We concentrate in this paper on the prominent cloud banding often found in the southern 76

quadrant of intense storms (Neiman et al. 1993; Grønås 1995; Browning and Field 2004), where

our observations reveal a relation between the wind strength and the cloud bands. We show that 78

a similar relation appears in an experimental trial of the Met Office high-resolution ensemble

forecast system. 80

THE DIAMET PROJECT. The DIAMET project is one of the three components of the UK Natural 82

Environment Research Council’s Storm Risk Mitigation Programme

(http://www.bgs.ac.uk/stormrm/home.html), and is a collaboration between British academic 84

groups and the Met Office. The main elements of the project, together with results published so

far, can be summarised as: 86

a) Use of field measurements and detailed numerical modelling to improve understanding of

key processes (Chagnon et al. 2013, Martínez-Alvarado et al. 2014, Norris et al. 2014) 88

b) Critical assessments of the performance of parameterisations of convection (Martínez-

Alvarado and Plant 2013), air-sea fluxes (Cook and Renfrew 2014) and microphysics 90

(Dearden et al. 2014, Lloyd et al. 2014) in numerical weather prediction models and

c) Addressing fundamental aspects of predictability using ensembles, and improving data 92

assimilation methods for better short-term forecasting (Baker et al. 2014)

94

The scientific approach of DIAMET concentrates on the effect of diabatic processes on the

distribution of potential vorticity (PV) and its consequences for the evolution of weather systems. 96

PV combines the vertical stability of the atmosphere with the horizontal shear and rotation of the

wind field, and is materially conserved in the absence of diabatic and frictional processes. It is a 98

local measure of circulation about a point and its distribution is fundamental to our understanding

5

of Rossby waves and the evolution of cyclones (e.g. Hoskins et al. 1985). Cyclone development 100

typically arises through interaction between a Rossby wave on the tropopause and large-scale

horizontal waves in temperature near the ground, with the surface cyclone center positioned to 102

the east of the upper-level PV maximum (trough). This process is well understood, but the effects

of PV anomalies produced by diabatic processes are much less clear. It is possible to attribute 104

diabatically-generated PV anomalies to the processes that produce them in a forecast model (e.g.,

Chagnon et al. 2013), but these diagnostics need to be tested by comparison with observations. A 106

main goal of DIAMET is to use detailed measurements of dynamics, cloud physics and air-sea

fluxes to calculate diabatic heating rates in cyclones and thereby evaluate how well the diabatic 108

production and removal of PV anomalies are represented in models. Key scientific questions are:

a) How do latent heating and air-sea fluxes modify the mesoscale potential vorticity 110

distribution in an extratropical cyclone?

b) How do such modifications affect the precipitation and wind fields and therefore the 112

impact of the cyclone on society?

c) Do these modifications affect the overall development of the cyclone (and others 114

downstream) or are they only important locally?

d) How sensitive are numerical forecasts to the parameterisation of latent heating and air-116

sea fluxes? Can model error from these sources be quantified?

From September 2011 to August 2012, DIAMET conducted four field campaigns lasting several 118

weeks each to examine cyclones around the UK and Ireland. The primary measurement platform

was the BAe146 aircraft of the UK’s Facility for Airborne Atmospheric Measurements (FAAM, 120

http://www.faam.ac.uk/) carrying instrumentation to measure winds, thermodynamic

parameters, microphysics and chemical tracers. A summary of the relevant instrumentation for 122

DIAMET is shown in Table 1. The FAAM aircraft can operate up to an altitude of 10 km with an

6

endurance of around 5 hours (see Renfrew et al. 2008 for more details and flight examples). In 124

several DIAMET cases, double flights were conducted, with a break for refuelling mid-mission. For

three of the four campaigns, the aircraft was based in Cranfield, north of London, but in the winter 126

campaign of November–December 2011 it was based at Exeter, near to the Met Office

headquarters (see Fig. 5 for locations described in the text). Ground-based measurements were 128

also a crucial part of DIAMET: 3D precipitation-radar measurements from the Chilbolton Facility

for Atmospheric and Radio Research (Browning et al. 2007); precipitation maps from the Met 130

Office’s operational precipitation-radar network (Harrison et al. 2009); continuous measurements

from wind profilers, especially the UK Mesosphere–Stratosphere–Troposphere VHF profiler at 132

Aberystwyth in Wales (Vaughan 2002); and surface data from the Met Office’s network of around

270 automatic weather stations (AWS). Fifty-five additional radiosonde launches were made from 134

selected stations on intensive observation period (IOP) days.

136

There were fifteen flying days (Table 2) covering fourteen DIAMET IOPs (flights into the same

meteorological system on successive days are grouped into the same IOP). Nine of these IOPs 138

were in 2011 and five in 2012, and all but one involved the FAAM aircraft. The exception, IOP 10,

was declared a DIAMET IOP because of the passage of a mesoscale convective system over the 140

Chilbolton radar which led to widespread flooding in the Severn valley, the largest river catchment

in the UK. Three flights are also listed from the T-NAWDEX (THORPEX-North Atlantic Waveguide 142

and Downstream impact Experiment) Pilot campaign in November 2009. They were conducted by

the DIAMET team as a preliminary investigation, and have been analysed as part of the DIAMET 144

project (labelled TNP1-3 in Table 2).

146

7

THE NORTH ATLANTIC WEATHER REGIME OF EARLY WINTER 2011–2012. The second DIAMET 148

aircraft campaign (24 November to 14 December 2011) was characterized by an unusually strong

zonal jet stream across the North Atlantic and a rapid succession of intense cyclones, many of 150

which crossed northwest Europe. One way to characterize the large-scale weather pattern over

the North Atlantic is to examine the North Atlantic Oscillation (NAO) index, a normalized measure 152

of the pressure difference between Iceland and the Azores. Positive values of the NAO are

associated with stormy weather over northwest Europe, with milder temperatures and greater 154

precipitation than the negative phase. Daily NAO index values are provided by the

NOAA/NCEP/Climate Prediction Center for January 1950 to June 2012 156

(http://www.cpc.ncep.noaa.gov/products/precip/CWlink/pna/nao.shtml). Remarkably, 30

November, 3 December, and 6 December 2011—all within the DIAMET campaign period—appear 158

in the top 0.4% of daily values of this long time series (22,825 days). December 2011 had the

highest monthly NAO index (2.52) of any December and the third highest for all months of the 160

entire time series.

The zonal wind strength between 1 November 2011 and 31 January 2012 is shown in Fig. 1a 162

together with the climatological mean and standard deviation for this time of year. The statistics

are calculated from the average of zonal wind at 300 hPa across the North Atlantic (between 40-164

60oN and 10-60oW) obtained from 6-hourly ERA-Interim re-analysis data (Dee et al. 2011). The

climatology was obtained from the entire ERA-Interim dataset 1979–2010 by calculating the mean 166

and standard deviation of each calendar date across all years and then smoothing the resulting

series with a 7-day running mean. The zonal wind was exceptionally strong during the DIAMET 168

campaign with values more than one standard deviation above the climatological mean for most

of the period and more than two standard deviations for a few days. Comparing the histogram of 170

North Atlantic zonal wind compiled from data during the 21 days of the DIAMET campaign with

8

that during the three-month period November 2011 – January 2012 (Fig. 1b), and the 172

corresponding histogram from all Nov-Dec-Jan periods in the ERA-Interim dataset, further

underlines the anomalous conditions during this period. 174

Between 22 November and 10 January, the Free University of Berlin named 29 significant cyclones

affecting Europe as part of their Adopt-A-Vortex scheme, corresponding to one new storm 176

forming every 1.7 days. All but one of these storms passed over or near the UK or affected the UK

with their frontal systems. Many of the stronger storms are reflected by decreases in the average 178

zonal wind across the North Atlantic. These decreases occur as the Rossby waves at tropopause

level reach large amplitude and break, causing the jet stream to split, with weaker average flow in 180

the 40–60°N band. Decreases in jet stream strength (often leading to minima) were associated

with the intense cyclonic storms: 182

Zafer (1 December, DIAMET IOP6) was a small-scale intense low north of Scotland.

Friedhelm (7–8 December, IOP8) was the explosively deepening cyclone discussed in Section 4. 184

Hergen (12-13 December, IOP9) passed north of Scotland extending an active warm front

across the UK. 186

Joachim (15–16 December) was fast moving and tracked directly into central Europe causing

widespread gales and heavy rain associated with some damage in southern Germany. 188

Patrick (25 December) and Robert (28 December) followed similar tracks across northern

Scotland. 190

Ulli and Andrea (3–5 January) again brought high winds to Scotland and northern England.

In contrast, the dips to relatively low values on 18–19 December 2011 and around 12 January 192

2012 were associated with ridge-building not linked to a preceding cyclone.

194

We now present a more detailed study of cyclone Friedhelm, the subject of DIAMET IOP8.

9

196

DIAMET IOP8 – FLIGHT INTO THE CENTER OF CYCLONE FRIEDHELM. The passage of extratropical

cyclone Friedhelm resulted in considerable disruption to transport and some damage to 198

infrastructure over the central belt of Scotland. Friedhelm started as a shallow wave on a trailing

cold front over Newfoundland at 1200 UTC on 6 December 2011. As it crossed the Atlantic, the 200

storm crossed the axis of the polar jet stream and deepened spectacularly—44 hPa between mid-

day on 7 December and mid-day on the 8th, easily qualifying as a meteorological bomb (Sanders 202

and Gyakum 1980). Structurally, Friedhelm resembled a Shapiro-Keyser cyclone (Shapiro and

Keyser 1990) with a frontal fracture (weakening of the northern part of the cold front near the 204

warm front) followed by wrapping of cold air at low levels around the northern flank of a warmer

cyclone center. By the time Friedhelm reached western Scotland early on 8 December, the storm 206

was in its mature stage with a central pressure of 957 hPa (Fig. 2). A large separation between the

warm sector and the cyclone center showed that the cyclone had occluded (Schultz and Vaughan 208

2011), with a bent-back front corresponding to the mass of cloud curling westward around

Scotland (Fig. 3a). Southwest of the cyclone center, bands of low cloud extended eastwards 210

towards Scotland from beneath the mid-level cloud deck. Corresponding banding was observed in

light precipitation over the sea to the west of Scotland by the Met Office radar network (Fig. 3b). 212

These bands (labelled A–D) moved east-southeastward across central Scotland. The storm

continued to move eastward towards Scandinavia, with the strongest winds crossing to the east 214

side of Scotland by early evening. The precipitation bands crossed Scotland with the storm and

were especially prominent around 1800 UTC over central Scotland (Fig. 4a). The cyclone also 216

wrapped up further with the strongest winds moving into the southern and then southeastern

side of the low pressure center. 218

10

Figure 4b shows the maximum wind gusts measured on 10-m masts at selected AWS across the

northern UK during 8 December. The numbers are coloured by the time of maximum gust to 220

illustrate the eastward progression of the high winds associated with the storm. Maxima of

around 30 m s–1 (67 mph) occur over a wide area of Scotland, notably in the highly populated 222

region between Glasgow and Edinburgh. The previous day the Met Office had issued its first ever

red alert for winds (warning the public to take action), allowing precautions to be taken including 224

the closure of schools, roads and bridges in the threatened areas. Over some of the Scottish

mountains, much higher winds were recorded – the highest being 74 m s–1 (165 mph) reported on 226

the summit of Cairn Gorm (1237 m), around 100 km north-northwest of Leuchars.

The FAAM aircraft was tasked on this day with investigating the region of strongest winds to the 228

south and southwest of the cyclone center. Such winds are often associated with a cold conveyor

belt, a low-level airstream which wraps around a cyclone as it develops (Carlson 1980). In addition, 230

some Shapiro-Keyser cyclones develop sting jets – cores of very strong low-level winds associated

with descending airstreams ahead of the mid-level cloud head (Browning 2004; Clark et al. 2005). 232

Indeed, evidence for such features during the passage of Cyclone Ulli on 3 January 2012 was

presented by Smart and Browning (2013). There have been other aircraft flights through intense 234

extratropical cyclones; for example, Neiman and Shapiro (1993) and Neiman et al (1993)

presented observations of an extremely rapidly developing cyclone (ERICA IOP4) which had similar 236

bent-back frontal structure to Friedhelm and prominent rainbands in the same sector of the storm

(their Fig. 19). DIAMET was able to build on these results by measuring the wind structure 238

associated with precipitation bands together with their cloud microphysical properties.

The aircraft took off at 1048 UTC from Exeter, timed to intercept the cyclone center before it had 240

traversed Scotland. Baker et al. (2013) described the thrill of the flight and showed photographs

taken from within the cyclone center looking across the curving cloud bands, as well as the sea 242

11

state in the high-wind region. The first leg of the flight, at an altitude of 7 km, launched ten

dropsondes along the west coast of Scotland to measure the thermodynamic and wind structure 244

along a transect through the storm to the north of the cold front and reached the cyclone center

at 1234 UTC (Fig. 5). 246

Figure 6 shows relative humidity with respect to ice (RHi), potential temperature and horizontal 248

winds along the dropsonde section. The sloping temperature gradient, between 2 and 6 km in the

southern part of the section, separated dry air above it from the moister air below (northwest of 250

the cold front). The wind-speed cross section shows the upper-level jet stream between 54° and

55.5°N, exceeding 60 m s–1 at 6-km altitude above a layer of pronounced wind shear. 252

A second, weaker, temperature gradient is shown below 4 km, north of 57°N. Here, temperature

increased with latitude (Fig. 6), associated with the bent-back warm front which had wrapped 254

around to the south side of the cyclone. The pool of warmer air at low levels near the core of this

kind of cyclone is called a seclusion (Shapiro and Keyser 1990). The low-level wind maximum that 256

was the focus of this mission lay on the southern flank of this front. An L-shaped wind maximum

extended from 4 km down to 1.5 km at 56.3°N, with an extension northward to 57.1°N below 2.5 258

km; a low-level zonal-wind maximum is expected from approximate thermal wind balance with

the poleward increase in potential temperature. Martínez-Alvarado et al. (2014) found that the air 260

entering the strong wind region south of Friedhelm’s center comprised three airstreams with

distinct trajectory origins and observed tracer composition. Two airstreams were associated with 262

the cold conveyor belt – the air was cloudy and back-trajectories stayed at low levels wrapping

around the cyclone core – whereas the third resembled a sting jet which left the tip of the cloud 264

head to the west of the cyclone and descended towards the east-southeast. In Fig. 6, the strongest

winds, measured by sondes 5–7, coincided with relative humidity values exceeding 80% beneath 266

12

the sloping isentropes, suggesting that the cold conveyor belt dominated the low-level winds

along this section. 268

After executing the second dropsonde section to the southwest of the cyclone center (Fig. 5; see

Martínez-Alvarado et al. 2014 for cross-section), the aircraft descended to make in-situ 270

measurements at lower altitudes. Of particular interest for this paper are the low-level legs to the

west of Scotland at around 1500 UTC, through the region of strongest low-level winds, which are 272

discussed in the next section. Further low-level work was conducted after refuelling in Teesside,

by which time the strongest winds were located to the east of Scotland. Here the relative humidity 274

(with respect to ice) of the fastest-moving air was around 42%, suggesting descent of around 150

hPa from initial saturation. Turbulent mixing was intense at low levels on this flight – the turbulent 276

kinetic energy, calculated from the 32-Hz turbulence probe, was 7–10 m2 s−2 – and the aircraft

became coated in sea salt even when flying at 500 m above sea level. (Observations of turbulence 278

throughout the DIAMET experiment are reported in Cook and Renfrew 2014.) On this boundary-

layer leg (lasting 30 minutes), the gradient in potential temperature was almost uniform with 280

values decreasing towards the south, whereas wind speed increased to an average of 47 m s–1 at

the southern end (not shown). Therefore, the aircraft was able to make detailed measurements of 282

the wind field and thermodynamic variables in the region of maximum wind speed to the south of

the cyclone center, both west and east of Scotland. 284

BANDING IN CLOUD, PRECIPITATION, AND WINDS ON THE SOUTHERN FLANK OF FRIEDHELM.

We now turn to the question of how the cloud and precipitation bands (Fig. 3) were linked to the 286

severe surface winds (Fig. 4b), concentrating on the low-level legs measured around 1500 UTC as

the aircraft flew from Islay northwards towards Tiree in a region of banded precipitation within 288

the cold conveyor belt airstream of the cyclone (Fig. 7). The bands in Fig. 7 are numbered from

south to north and fall in two groups: the northern group (B) intersected by the aircraft and a 290

13

southern group (S) with a more counterclockwise orientation. As the progression from white lines

(position at 1500 UTC) to yellow lines (1515 UTC) shows, the bands were moving quickly south-292

southeastwards. The aircraft cross section crossed B0, B1 and B2, reaching its northern turn in a

gap in the precipitation along the much-larger B3 band which flanks the southern side of the bent-294

back front.

Figure 8a shows the relative humidity from the aircraft overlain with surface precipitation rate 296

estimated from the radar network (red). The flight leg crossed two cloud bands (RH ≥ 100%),

identified as bands B1 and B2 in Fig. 7, which were clearly precipitating at the time of crossing (no 298

cloud was encountered at aircraft altitude during the passage of B0). Note how the wind speed is

weaker in the bands than in the clearer air in between (Fig. 8b). The aircraft began its 180o 300

platform turn at 1516 UTC just as it reached B3, rendering its wind measurements unreliable, so

the section of flight through B3 is not shown in Fig. 8. 302

FIgures 8c and 8d show measurements of liquid and ice-number concentrations across the bands,

from the Cloud Droplet Probe (CDP) and Cloud Imaging Probe (CIP-100) respectively; details of the 304

microphysics instrumentation are given in Lloyd et al. (2014). The clouds contained mixed liquid

and ice, with relatively high concentrations of ice particles, images of which show complex 306

aggregates of plate-like crystals (Fig. 9a) as well as high concentrations of columnar crystals at

temperatures between -5° and -8°C, characteristic of secondary ice formed by rime splintering 308

(Hallett and Mossop 1974) (Fig. 9b). Substantial ice concentrations were also found in regions sub-

saturated with respect to ice. One of the main objectives of DIAMET was to quantify diabatic 310

heating and cooling rates using the observed microphysics, to compare with and improve model

simulations. As an example of this, the method described in Dearden et al (2014) has been used to 312

calculate instantaneous diabatic heating and cooling rates associated with the growth and

evaporation of ice crystals by vapor diffusion along this section of the flight (Fig 8e). The 314

14

calculations made use of relative humidity derived from the WVSS-II hygrometer data, and ice

particle size distributions from the CIP-100 probe in the range 0.1 to 6.2 mm (this probe agreed 316

well with other probes in the size range below 1 mm, see Fig 9c). We obtained information on ice

particle shapes from the imaging probes and used this to constrain the calculations (red line, Fig 318

8e). However, many existing bulk microphysics schemes assume fixed shapes for ice crystals to

determine the rate of change of ice mass by vapor diffusion, with some simply assuming spherical 320

particles such that the dimension of a particle represents its diameter (e.g. Wilson and Ballard

1999, as used by the Met Office Unified Model; Morrison et al. 2005; Liu et al. 2003). To explore 322

the impact of particle geometry, calculations were also performed assuming the ice particles were

spherical during diffusion growth, with a diameter equal to the maximum dimension (blue line, Fig 324

8e). The results show that for a given particle size distribution, the assumption of spherical shapes

leads to an overestimation of the heating and cooling rates by a factor of two or more. 326

The regions of diabatic heating at 1510–1511 UTC and 1513–1514 UTC both coincide with small

decreases in wind speed, consistent with transport of lower-momentum air from below in shallow 328

convective updrafts. The instantaneous cooling rates in the clear air were observed on the

southern side of the cloud bands. This raises a question regarding the dynamical mechanism 330

responsible for the emergence of banding in this region of intense cyclones that could explain the

relationship between the cloud and wind in such bands, and whether there could be a cooperative 332

feedback from diabatic heating and cooling on the winds (both vertical and horizontal

components). We now examine surface wind observations during the passage of the bands. 334

SURFACE WIND OBSERVATIONS. Figure 10 shows the position of rainbands with time along a

straight line joining AWS sites on the islands of Tiree and Islay (Fig. 8). Distance is measured from 336

north-northwest to south-southeast. The precipitation rate at 5-minute intervals estimated by the

radar network has been interpolated to the section to create the time–distance progression of 338

15

rainfall rate. Wind-gust strength (5-minute running median to be consistent with the radar update

interval) is overlain for both AWS sites, after subtracting a 90-minute running median to remove 340

the larger-scale variation. White curves indicate the progression of rainbands along the section,

identified using animations of the radar images. Band B1 moved from Tiree (point T) at 1445 UTC 342

to Islay (I) at 1610 UTC and was intercepted by the aircraft at point A (1510 UTC). Note the

variability in the precipitation rate of this rainband as it moved along the section; as well as 344

moving south-southeastward, the cloud bands changed morphology noticeably between 5-minute

radar images. One rainband appeared to split into two at around 1500 UTC: band B1 moved 346

towards the ESE whereas B2 was more stationary, curving almost parallel to the bent-back front

(Fig. 7). The rain reaching Tiree at 1515 UTC (Fig. 7) was at the tip of a much broader precipitation 348

feature advancing rapidly from the west-northwest along the bent-back front, which we do not

discuss here. At about 1615 UTC, another precipitation band B4 emerged between band B2 and 350

the bent-back front.

Consistent with the aircraft observations at 840 hPa, the surface wind speed was lower in the core 352

of rainband B1 and immediately after it (at T and I) than in the clear air either side. Similar dips in

wind speed occurred during the passage of band B2 across Islay (1715 UTC) and also the earlier 354

bands S1 and S2. To see if this relationship held more generally, data from all 13 AWS sites across

central Scotland were examined for this day. The mode of wind speed within the rainbands was 356

1.5 m s–1 lower than between them, while locally the peak-to-trough variation in surface gust

strength associated with bands was as much as 10 m s-1. In summary, both at aircraft altitude and 358

at the surface there is evidence that the wind speed tends to dip during the passage of a rainband,

and to be higher in the clear air in between. Given the linear nature of the rainbands and their 360

tendency to align along the mean wind, this suggests that the strongest surface winds (and

16

therefore damage) will be arranged in linear swathes. We now examine forecast model 362

predictions of the storm to see how well the Met Office Unified Model simulated these features.

ENSEMBLE PREDICTION OF BANDING IN THE HIGH-WIND REGION. DIAMET IOP8 provided an 364

ideal test case for the new MOGREPS (Met Office Global and Regional Ensemble Prediction

System) convection-permitting forecast system prior to its operational implementation. The 366

MOGREPS-UK ensemble is now run routinely and consists of 12 forecasts run every 6 h with the

Met Office Unified Model, on a limited area spanning the UK with a horizontal grid spacing of 2.2 368

km. For DIAMET IOP8, each ensemble member took its initial and boundary conditions from the

corresponding North Atlantic and European regional version of MOGREPS (Bowler et al. 2008). 370

Figure 11 shows a snapshot at 1600 UTC from the first four MOGREPS-UK members, zooming in on

Scotland. The color shading is the 850-hPa wind speed, and lines indicate the axes of wind-speed 372

maxima. Figure 12 overlays the same lines on the model precipitation rates. The high-wind cores

generally lie along the clear slots between the rainbands. The same is true of the operational 374

deterministic Met Office 1.5-km (UKV) model forecast (not shown). This structure and magnitude

of variation is consistent with the in-situ aircraft observations presented in Fig. 8. 376

Each of the 12 ensemble forecasts exhibited banding in precipitation and winds, and we now

examine how well the observed precipitation structure was represented and its dependence on 378

length scale (L). Roberts and Lean (2008) and Roberts (2008) introduced a measure of the

similarity of a forecast pattern of precipitation rate to the radar-derived rate called the Fractions 380

Skill Score (FSS). First, the radar and forecast data are interpolated onto the same regular grid.

Then neighbourhoods are defined as squares of side L and we compute the fraction of pixels 382

within each neighbourhood that have a rain rate exceeding a certain threshold. The calculation is

repeated for neighbourhoods centered on every grid box, using both radar and forecast data. If 384

the precipitation fraction in a forecast matches the observed fraction in every neighbourhood

17

square, the FSS equals 1. The lowest possible score for a complete mismatch everywhere is 0, and 386

a score of 0.5 indicates a minimum level of satisfactory spatial agreement (the forecast pattern is

in agreement more often than it is not). 388

Figure 13 shows the FSS versus scale, averaged for Scotland over forecast lead times of 4–24 hours

(for a rain rate threshold of 2 mm hr–1). The FSS increases with scale and exceeds 0.5 for scales 390

greater than 25 km for the best ensemble member. The precipitation pattern is highly dependent

upon initial conditions, with the worst ensemble member having FSS < 0.5 even for the 110-km 392

scale (associated with a displacement of the cyclone). The precipitation pattern is dominated by

the bands which have a characteristic spacing ranging from 20 to 50 km, so the statistics indicate 394

that the model must have captured the positioning of the bands to some extent. Those ensemble

members that have a strong match (high FSS) when the bands first emerge maintain the highest 396

FSS throughout the forecast, showing that the model is capable of simulating the evolution of the

bands, given favourable initial conditions. Currently, the DIAMET team is evaluating the 398

MOGREPS-UK ensemble performance in precipitation forecasts using FSS, and relating skill in

precipitation to the skill in the forecasts of mesoscale features which are associated with 400

precipitation, such as fronts.

CONCLUSIONS. The second DIAMET campaign took place at a time of exceptionally strong flow 402

across the North Atlantic and vigorous cyclone frequency and intensity over northwestern Europe.

The DIAMET flights were the first aircraft missions with comprehensive cloud instrumentation to 404

sample the strong wind region south of a cyclone center. IOP8 took the investigators into the

extreme winds of extratropical cyclone Friedhelm, which had a T-bone frontal structure with bent-406

back front. A phenomenon of particular interest was the cloud and precipitation banding in the

strong winds found in the southern quadrant of the storm. 408

18

The FAAM aircraft traversed this region between 2000 and 700 m altitude crossing three cloud

bands, two of which were precipitating at the time. The clouds were found to be of mixed phase 410

with high number concentrations of secondary ice crystals in the form of columns, consistent with

ice multiplication by the Hallett–Mossop (splintering) process. Estimates of the diabatic heating 412

rate were found to be sensitive to the particle shape: assuming spherical ice particles when

calculating diffusional growth can lead to errors of a factor of two in the heating rate. Between the 414

bands, sublimating ice particles gave instantaneous diabatic cooling rates of up to 4 K hr–1 in a

region where (according to dropsonde 4) the atmosphere was close to neutral stability below 800 416

hPa (2 km).

The horizontal wind speed was lower within the cloud bands than in the clear air between. At 840 418

hPa, diabatic heating (deduced from the microphysical measurements) occurred where the wind

speed decreased, indicating that saturated updrafts were transporting lower-momentum air from 420

the boundary layer below. However, wind measurements from automatic weather stations at the

ground also showed a dip in wind speed as rainbands passed overhead. Thus, a link existed 422

between precipitation banding and wind speed throughout the lowest 2 km. Further work is under

way to understand the dynamical reasons for the bands and their characteristic spacing. 424

The DIAMET IOP8 case was one of several used to examine the behaviour of the Met Office

MOGREPS-UK 2.2-km ensemble prior to its operational implementation. The forecasts indicated 426

banding in winds and precipitation with the highest wind speeds in the clear slots. The best

member matched the structure in precipitation rates observed by radar for length scales of 25 km 428

and above, and this degree of matching was sustained throughout the forecast. This length scale

corresponded with the rainband spacing, indicating that the model was capable of forecasting the 430

rainbands and associated wind structures with a positional uncertainty similar to the spacing

19

between the bands. However, this is only one case, and ensemble forecasts from many cases are 432

required to quantify the skill in probability forecasts of mesoscale features.

Cloud and precipitation banding is often observed in satellite and radar imagery in the southern 434

quadrant of intense extratropical storms. The finding that the precipitation bands are associated

with structure in the wind with gusts up to a few m s–1 higher in the cloud-free slots, even at the 436

surface, is important for predicting the local impact of cyclonic windstorms. For example, in simple

environmental risk models wind damage scales with the third power of wind speed, so even small 438

wind-speed enhancements of several m s–1 are potentially important. In IOP8, the rainbands were

aligned approximately with the large-scale wind direction, so the highest wind gusts would have 440

been concentrated along linear swathes. This coherent structure of the wind fields is important

for nowcasting, as well as for short-range forecasting where there is now potential skill in 442

predicting wind-band structure (even if not the precise location) using high-resolution models.

Acknowledgments. DIAMET is funded by the Natural Environment Research Council (NERC) as 444

part of the Storm Risk Mitigation Programme (NE/I005234/1) in collaboration with the Met Office.

The BAe-146 Atmospheric Research Aircraft is flown by Directflight Ltd. and managed by the 446

Facility for Airborne Atmospheric Measurements (FAAM) on behalf of NERC and the Met Office.

We particularly thank Captain Alan Foster and Co-Pilot Ian Ramsay-Rae for flying in such extreme 448

conditions and working hard to find an airport that would let us land for refuelling during the

storm. Figure 3a was supplied by the Dundee Satellite Receiving Station. PK, JO, and TT received 450

funding from the AXA Research Fund as part of the Seamless Approach to Assessing Model

Uncertainties in Climate Projections of Severe European Windstorms (SEAMSEW) project. 452

20

Box 1: Educational resources

A research project on storms presents an excellent opportunity for outreach activities, and in 454

DIAMET we collaborated with an award-winning educational consultant, Heather Reid (a former

broadcast meteorologist), to develop a package of educational resources for school children 456

related to our key science aims. These included two short professionally-filmed videos on

“Forecasting the weather” and “Studying severe storms around the UK,” as well as four sets of 458

exercises comprising information for the pupils and activities or worksheets. Topics covered

included change of state, latent heat, the electromagnetic spectrum, and using observational data. 460

The videos and worksheets are available on the NCAS project website,

http://www.ncas.ac.uk/index.php/en/diamet-schools, and are being actively promoted via project 462

partners like the Royal Meteorological Society, the Institute of Physics, and Education Scotland.

The material is aimed at students aged 12–14 years taking science, in particular physics. The topics 464

were chosen to fit broadly into the secondary-school physics curriculum and provide alternative

contexts for the “uses of physics” to the more common examples like X-rays in medical physics or 466

heating solid stearic acid to demonstrate latent heat. This topic choice was deliberate. First, the

topics fit neatly with the project’s focus on diabatic processes and the use of a research aircraft. 468

Second, all too often weather and climate feature in geography rather than physics at school level;

a physics background can be vital to access university teaching of these subjects and, more 470

importantly, to develop the skills usually required for a career in the field.

472

21

References 474

Baker L. H., O. Martínez-Alvarado, J. Methven, and P. Knippertz, 2013: Flying through extratropical

cyclone Friedhelm. Weather, 68, 9–11. 476

Baker, L., Rudd, A., Migliorini, S. and Bannister, R., 2014: Representation of model error in a

convective-scale ensemble prediction system. Nonlinear Processes in Geophysics, 21. 19-39. 478

Bowler, N. E., A. Arribas, K. Mylne, K. B. Robertson and S. E. Beare, 2008: The MOGREPS short-

range ensemble prediction system, Quart. J. Roy. Meteor. Soc., 134, 703-722. 480

Browning, K., 2004: The sting at the end of the tail: Damaging winds associated with extratropical

cyclones. Quart. J. Roy. Meteorol. Soc., 130, 375–399. 482

Browning, K. A. and M. Field, 2004: Evidence from Meteosat imagery of the interaction of sting 484

jets with the boundary layer. Meteorol. Appl., 11, 277–289.

486

Browning, K. A., A. M. Blyth, P. Clark, U. Corsmeier, C. Morcrette, J. Agnew, D. Bamber, C.

Barthlott, L. Bennett, K. Beswick, M. Bitter, K. Bozier, B. Brooks, C. Collier, C. Cook, F. Davies, B. 488

Deny, M. Engelhardt, T. Feuerle, R. Forbes, C. Gaffard, M. Gray, R. Hankers, T. Hewison, R.

Huckle, N. Kalthoff, S. Khodayar, M. Kohler, S. Kraut, M. Kunz, D. Ladd, J. Lenfant, J. Marsham, J. 490

McGregor, J. Nicol, E. Norton, D. Parker, F. Perry, M. Ramatschi, H. Ricketts, N. Roberts, A. Russell,

H. Schulz, E. Slack, G. Vaughan, J. Waight, D. Wareing, R. Watson, A. Webb, A. Wieser, and K. Zink, 492

2007: The Convective Storm Initiation project. Bull Amer. Met. Soc., 88, 1939–1955.

494

Carlson, T. N., 1980: Airflow through midlatitude cyclones and the comma cloud pattern. Mon.

Wea. Rev., 108, 1498–1509. 496

22

Chagnon, J., S. L. Gray, and J. Methven, 2013: Diabatic processes modifying potential vorticity in a 498

North Atlantic cyclone. Quart. J. Roy. Meteorol. Soc., 139, 1270–1282.

500

Clark, P. A., K. A. Browning, and C. Wang, 2005: The sting at the end of the tail: Model diagnostics

of fine-scale 3-D structure of the cloud head. Quart. J. Roy. Meteorol. Soc., 131, 2263–2292. 502

Cook, P. A. and I. A. Renfrew, 2014: Aircraft-based observations of air-sea turbulent fluxes 504

around the British Isles. Quart. J. Roy. Meteorol. Soc., DOI: 10.1002/qj.2345, in press.

506

Dearden, C., P. J. Connolly, T. W. Choularton, G. Lloyd, K. Bower, J. Crosier, and G. Vaughan, 2014:

Diabatic heating and cooling rates from in-situ microphysics measurements: A case study of a 508

wintertime UK cold front. Mon. Wea. Rev., DOI: 10.1175/MWR-D-14-00048.1, in press.

510

Dee, D. P., S. M. Uppala, A. J. Simmons, P. Berrisford, P. Poli, S. Kobayashi, U. Andrae, M. A.

Balmaseda, G. Balsamo, P. Bauer, P. Bechtold, A. C. M. Beljaars, L. van de Berg, J. Bidlot, N. 512

Bormann, C. Delsol, R. Dragani, M. Fuentes, A. J. Geer, L. Haimberger, S. B. Healy, H. Hersbach, E.

V. Holm, L. Isaksen, P. Kallberg, M. Koehler, M. Matricardi, A. P. Mcnally, B. M. Monge-sanz, J. -. 514

Morcrette, B. -. Park, C. Peubey, P. de Rosnay, C. Tavolato, J. -. Thepaut, and F. Vitart, 2011: The

Era-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. 516

Roy. Meteorol. Soc., 137, 553–597.

518

Grønås, S., 1995: The seclusion intensification of the New Year's Day storm 1992. Tellus, 47A, 733-

746. 520

23

Hallet, J., and S. C. Mossop, 1974: Production of secondary ice particles during the riming process. 522

Nature, 249, 26–28.

524

Harrison, D. L., R.W. Scovell, and M. Kitchen, 2009: High-resolution precipitation estimates for

hydrological uses. Proceedings of the ICE - Water Management, 162, 125–135. 526

Hoskins, B. J., M. E. McIntyre, and A. W. Robertson, 1985: On the use and significance of isentropic 528

potential vorticity maps. Quart. J. Roy. Meteor. Soc., 111, 887–946.

530

Liu, H-C., P. K. Wang and R. E. Schlesinger, 2003: A numerical study of cirrus clouds. Part I: Model

description. J. Atmos. Sci., 60, 1075-1084. 532

Lloyd, G., C. Dearden., T. W. Choularton., J. Crosier and K. N. Bower, 2014: Microphysics 534

observations of warm, cold and occluded frontal systems during the DIAMET campaign. Mon.

Wea. Rev., submitted. 536

Martínez-Alvarado, O. and R. S. Plant, 2013: Parameterised diabatic processes in numerical 538

simulations of an extratropical cyclone. Quart. J. Roy. Meteorol. Soc., doi: 10.1002/qj.2254, in

press. 540

Martínez-Alvarado, O., L. H. Baker, S. L. Gray, J. Methven, and R. S. Plant, 2014: Distinguishing the 542

cold conveyor belt and sting jet air streams in an intense extratropical cyclone. Mon. Wea. Rev.,

doi:10.1175/MWR-D-13-00348.1, in press. 544

24

Morrison, H., J. A. Curry and V. I. Khvorostyanov, 2005: A new double-moment microphysics 546

parameterization for application in cloud and climate models. Part I: Description. J. Atmos. Sci., 62,

1665-1677. 548

Neiman, P. J., and M. A. Shapiro, 1993: The life cycle of an extratropical marine cyclone. Part I: 550

Frontal-cyclone evolution and thermodynamic air–sea interaction. Mon. Wea. Rev., 121, 2153–

2176. 552

Neiman, P. J., M. A. Shapiro and L. S. Fedor, 1993: The life cycle of an extratropical marine cyclone. 554

Part II: Mesoscale structure and diagnostics. Mon. Wea. Rev., 121, 2177–2199.

556

Norris, J., G. Vaughan and D. M. Schultz, 2014: Precipitation in idealised baroclinic waves. Mon.

Wea. Rev., doi: 10.1175/MWR-D-13-00343.1, in press. 558

Renfrew, I. A., G. W. K. Moore, J. E. Kristjánsson, H. Ólafsson, S. L. Gray, G. N. Petersen, K. Bovis, P. 560

R. A. Brown, I. Føre, T. Haine, C. Hay, E. A. Irvine, A. Lawrence, T. Ohigashi, S. Outten, R. S. Pickart,

M. Shapiro, D. Sproson, R. Swinbank, A. Woolley, and S. Zhang, 2008: The Greenland Flow 562

Distortion Experiment. Bull. Amer. Meteor. Soc., 89, 1307–1324.

564

Roberts, N., 2008: Assessing the spatial and temporal variation in the skill of precipitation

forecasts from an NWP model. Met. Apps., 15, 163–169. 566

Roberts, N., and H. W. Lean, 2008: Scale-selective verification of rainfall accumulations from high-568

resolution forecasts of convective events. Mon Wea. Rev., 136, 78–97.

570

25

Sanders, F., and J. R. Gyakum, 1980: Synoptic-dynamic climatology of the “bomb.” Mon. Wea.

Rev., 108, 1589–1606. 572

Schultz, D. M., and G. Vaughan, 2011: Occluded fronts and the occlusion process: A fresh look at 574

conventional wisdom. Bull. Amer. Meteor. Soc., 92, 443–466, ES19–ES20.

576

Shapiro, M. A., and D. A. Keyser, 1990: Fronts, jet streams and the tropopause. Extratropical

Cyclones, The Erik Palmén Memorial Volume, C. W. Newton and E. O. Holopainen, Eds., Amer. 578

Meteor. Soc., 167–191.

580

Smart, D. J., and K. A. Browning, 2013: Attribution of strong winds to a cold conveyor belt and

sting jet. Quart. J. Roy Meteorol. Soc., doi: 10.1002/qj.2162. 582

Vaughan, G., 2002: The UK MST radar. Weather, 57, 67–73. 584

Wilson, D. R., and S. P. Ballard, 1999: A microphysically based precipitation scheme for the UK 586

Meteorological Office Unified Model. Quart. J. Roy. Meteor. Soc., 125, 1607–1636.

588

26

Figure Captions

Figure 1. Evolution of jet-stream strength indicated by the zonal wind at 300 hPa averaged over 590

the North Atlantic. (a) Time series from 1 November to 31 January showing 2011–2012 values in

blue from ERA-Interim, with the climatological mean (1979–2010) and standard deviation in black 592

(smoothed with a running 7-day mean). The DIAMET campaign period is marked. The letters refer

to the strongest cyclones passing over the UK (see text). (b) Normalised histogram of zonal wind 594

for the DIAMET campaign (red), November 2011 to January 2012 (blue), and Nov–Dec–Jan for the

whole ERA-Interim period 1979–2010 (black), estimated from 6-h data using Gaussian kernel 596

smoothing. The area under each curve integrates to 1.

Figure 2. Met Office surface analysis for 1200 UTC 8 December 2011 (Crown copyright). 598

Figure 3. (a) Infrared image from the AVHRR instrument on NOAA-19 at 1235 UTC 8 December

2011. (b) Rain rate (mm hr–1) at 1300 UTC estimated by the Met Office radar network (1-km grid 600

spacing). At 1234 UTC, the FAAM aircraft reached the storm center (pink dot). A–D indicate

rainbands moving to the east-southeast. 602

Figure 4. (a) Radar-derived precipitation rate (mm hr–1) at 1800 UTC 8 December 2011 when the

cyclone center had crossed to northeast Scotland and the banding to the south was most 604

prominent. (b) Maximum 1-minute gusts at surface stations over central Scotland during 8

December 2011, filtered using a 10-minute median. Gust strength (ms–1) is colored by the time of 606

occurrence according to scale in upper left corner of the panel.

Figure 5. Path of the FAAM aircraft on 8 December 2011 with the track colored according to 608

altitude. Black dots indicate dropsonde launches. The flight took off from Exeter at 1048 UTC,

landed for refuelling in Teesside at 1607 UTC, took off again at 1729 UTC, and returned to Exeter 610

27

at 2110 UTC. The aircraft was at low levels within the strongest winds at around 1500 UTC and

again at 1900 UTC. 612

Figure 6. (Left panel) Cross sections derived from the ten dropsondes released along the first flight

leg from 1130 to 1234 UTC of: left panel, relative humidity with respect to ice (colors) and 614

potential temperature (white contours); right panel, wind speed (ms–1, colors ) and wind direction

(barbs); the wind barbs use the usual convention for wind strength in knots. Numbers in green 616

denote the order in which sondes were dropped and are placed at the corresponding latitude. In-

situ measurements from the aircraft, flying at a constant pressure of 390 hPa, are shown in the 618

strip at the top.

Figure 7. Precipitation-radar image for 1515 UTC showing the bands intercepted by the aircraft 620

between 1505 and 1516 UTC (rain rate mm hr–1, colored according to scale at bottom). White

lines: bands at 1500 UTC corresponding to the labels in Fig. 8a (B3 from shape of clouds where 622

precipitation is absent). Yellow lines: positions of the bands on the 1515 UTC image, showing the

southeastward progression of the bands. The southern bands S1 and S2 were not intercepted by 624

the aircraft. White dots on Tiree and Islay denote positions of automatic weather stations.

Figure 8. Measurements from the FAAM aircraft as it flew northward from Islay to Tiree through 626

the strongest low-level winds. (a) Relative humidity with respect to ice along the flight track,

computed using WVSS-II data (shading) and rain rate (red line) derived from the radar network 628

(interpolated to the flight track). (b) Wind speed (m s–1, black; the winds highlighted in yellow

correspond to regions where the droplet number concentration exceeds 20 cm-3 or the ice 630

number concentration exceeds 5 L-1), radar altitude (x0.1km, grey dashed), temperature (°C, red)

and potential temperature (°C, blue). (c) Droplet number concentration (cm–3), as measured by 632

the Cloud Droplet Probe. (d) Ice particle number concentration (l–1) as measured by the CIP-100

probe. (e) Diabatic heating and cooling rates associated with ice deposition and sublimation of ice 634

28

crystals. Red line: mean value calculated over 8 s ( ~ 1 km) intervals using measured particle sizes

and shapes; grey shading: 1 Hz values, indicating the variability; blue line: as red line assuming 636

spherical particles and exponential size distribution. Wind speed is shown again in black and

yellow, with an expanded scale 638

Figure 9. (a) Images of ice crystals captured by the Cloud Particle Imager (CPI) within cloud band

B1 (from 1510 – 1511 UTC), showing a mixture of columns and complex plate aggregates; (b) as (a) 640

in the middle of cloud band B2 , revealing the dominance of columnar crystals, with evidence of

riming and aggregation; (c) comparison of particle size distributions (number of particles m-3 per 642

size bin, normalised to unit width) from the CIP-15 and CIP-100 probes, averaged over 1504–1516

UTC. The CIP-15 has a bin width of 15 μm and a maximum detectable size of 930 μm. The CIP-100 644

has a bin width of 100 μm and can detect precipitation-sized particles up to 6.2 mm.

Figure 10. A time–distance plot of radar-derived precipitation rate (mm hr–1, colored according to 646

scale at bottom) interpolated from the Met Office radar composite to a line connecting

observation sites at Tiree and Islay. Distance increases along the section from north-northwest to 648

south-southeast. Labels T and I identify the passage of rainband B1 over Tiree and Islay,

respectively, and point A indicates the crossing of this section by the aircraft. The time series of 650

wind gusts (black lines) measured at both AWS sites is overlain at the corresponding distance

along the section. A 90-minute running median has been removed from the winds to emphasise 652

the bands. The white curves indicate the progression of rainbands along the section (Fig. 8).

Figure 11. The 850-hPa wind speed (m s–1, colored according to scale at bottom) from the first four 654

members of the MOGREPS-UK trial forecast at 1600 UTC 8 December 2011, 7 h into the forecast.

Dashed lines indicate the axes of the wind maxima. 656

29

Figure 12. The rain rate (mm h–1, colored according to scale at bottom) from the first four

members of the MOGREPS-UK trial forecast at 1600 UTC 8 December 2011, 7 h into the forecast. 658

Red lines indicate the axes of the wind maxima identified in Fig. 11.

Figure 13. Fractions skill score (FSS) measuring the degree of fit between the rain rate pattern in 660

each forecast and the radar data versus horizontal scale, averaged over forecast lead times 4–24

h. The yellow shading indicates the full range of results from the 12-member ensemble and the 662

members 0, 4 and 5 are also indicated. The best ensemble member has a similar pattern to the

radar data (FSS > 0.5) for scales greater than 25 km. FSS is calculated over Scotland (54.6–59.3°N, 664

8.1°W–0.4°E).

666

668

670

672

30

Table 1: Instruments carried on the FAAM aircraft for DIAMET

Measurement Instrument Key parameters

Temperature Platinum resistance thermometer 32 Hz, ±0.3°C

Water Vapor General Eastern 1011B (-25 - +50°C)

Buck CR2 (-60 - +30°C) Spectra Sensors WVSS–II

tuneable diode laser

4 Hz, ±0.1 - 1 °C in dew/frost pt

1 Hz, ±0.1 - 0.5°C 0.4 Hz

Winds and turbulence FAAM 5-hole probe 32 Hz, 0.25 m s-1

Profiles below the aircraft Pressure, temperature,

humidity Winds Cloud top

Vaisala AVAPS RD94 dropsondes GPS tracking of dropsonde Leosphere ALS450 backscatter

lidar

2 Hz, ±0.4 hPa, ±0.2°C, ±2%

RH 4 Hz 5 - 30 s (along-track) , 1.5 m

(vertical resolution)

Liquid Water Johnson-Williams hot wire Nevzorov total water probe

4 Hz, ±0.3 g m-3 8 Hz, ±10%

Cloud and aerosol particles DMT CIP-15 imaging probe DMT CIP-100 imaging probe DMT CDP scattering probe DMT Cloud, Aerosol and

Precipitation Spectrometer with Depolarisation (CAPS-DPOL)

SPEC 2D-S shadow probe SPEC CPI V1.5 imaging probe DMT Passive Cavity Aerosol

Spectrometer Probe (PCASP)

1 Hz, 15 < D < 930 µm 1 Hz, 100 < D < 6200 µm 10 Hz, 3 < D < 50 µm 1 Hz, 15 < D < 1000 µm 100 Hz, 10 < D < 1280 µm 40 Hz, 5 < D < 1000 µm 1 Hz, 0.6 < D < 50 µm

Chemical species Ozone Carbon Monoxide Greenhouse Gases

TECO 49C UV analyser Aerolaser AL5002 fluorescence Los Gatos Cavity Enhanced

Absorption FGGA CO2 CH4

10 – 30 s, ±2 ppbv 1 Hz, ±4 ppbv 1 Hz, ±0.17 ppmv 1 Hz, ±1.3 ppbv

Upwelling infrared radiation Heimann KT-19.82 sensor ARIES Fourier Transform

Spectrometer

1 Hz, ±0.3 K brightness temperature

4 Hz, 3 – 18 µm, ±0.2 K brightness temperature

674

676

31

Table 2 DIAMET and T-NAWDEX Pilot Intensive Observation Periods

IOP Date Flight Duration, hours

Drop sondes

Scientific Objective

IOP1 16-Sep-11 B647 4.67 9 Convective rainband ahead of upper-level trough

IOP2 20-Sep-11 B648 (D) 7.35 15 Mesoscale waves running along trailing cold front – good coverage from Chilbolton

IOP3 23-Sep-11 B650 (D) 7.52 18 Rainband developing in diabatic Rossby wave beneath a warm conveyor belt

IOP4 26-Nov-11 B652 5.12 1 Surface fluxes in cold airstream approaching Scotland from the northwest

IOP5a 28-Nov-11 B654 4.75 15 Dropsonde profile across double front approaching from the Atlantic

IOP5b 29-Nov-11 B655 (D) 7.03 13 Intense cold front crossing UK from the west, giving rise to tornadoes on landfall

IOP6 01-Dec-11 B656 5.40 10 Small-scale cyclone Zafer near Shetland; measuring surface fluxes in high winds

IOP7 05-Dec-11 B657 3.13 0 Organised convection west of Scotland

IOP8 08-Dec-11 B658 (D) 9.00 21 Severe winter cyclone Friedhelm; Sting jet case

IOP9 12-Dec-11 B662 4.83 17 Warm front approaching from the west, bringing South coast gales and rainfall

IOP10 30-Apr-12 Slow moving cyclone bringing floods; overnight observations from Chilbolton radar

IOP11a 09-May-12 B694 4.60 8 Warm front of a frontal wave cyclone approaching from the southwest

IOP11b 10-May-12 B695 5.15 19 Warm front of same frontal cyclone over Scotland plus surface fluxes

IOP12 10-Jul-12 B712 4.27 9 Convective rainbands north of a mesoscale PV anomaly

IOP13 18-Jul-12 B715 4.60 11 Stationary warm conveyor belt over Scotland, bringing flooding

IOP14 15-Aug-12 B728 4.50 8 Bent-back front of strong summer cyclone over Ireland

TNP1 03-Nov-09 B483 4.65 11 Cold front capped by tropopause fold. Later developed tornadoes across S. England

TNP2 13-Nov-09 B486 4.70 17 Warm front at leading edge of frontal wave cyclone

TNP3 24-Nov-09 B488 5.23 7 Circuit around surface cold front over ocean and survey of warm conveyor belt

Note: flights marked (D) were double flights where the aircraft landed for refuelling mid-mission.678

32

Figures

680

682

Figure 1. Evolution of jet-stream strength indicated by the zonal wind at 300 hPa averaged over the North Atlantic. (a) Time series from 1 November to 31 January showing 2011–2012 values in 684

blue from ERA-Interim, with the climatological mean (1979–2010) and standard deviation in black (smoothed with a running 7-day mean). The DIAMET campaign period is marked. The letters refer 686

to the strongest cyclones passing over the UK (see text). (b) Normalised histogram of zonal wind for the DIAMET campaign (red), November 2011 to January 2012 (blue), and Nov–Dec–Jan for the 688

whole ERA-Interim period 1979–2010 (black), estimated from 6-h data using Gaussian kernel smoothing. The area under each curve integrates to 1. 690

692

33

694

Figure 2. Met Office surface analysis for 1200 UTC 8 December 2011 (Crown copyright). 696

698

700

34

Figure 3. (a) Infrared image from the AVHRR instrument on NOAA-19 at 1235 UTC 8 December 2011. (b) Rain rate (mm hr–1) at 1300 UTC estimated by the Met Office radar network (1-km grid 702

spacing). At 1234 UTC, the FAAM aircraft reached the storm center (pink dot). A–D indicate rainbands moving to the east-southeast. 704

706

35

708

Figure 4. (a) Radar-derived precipitation rate (mm hr–1) at 1800 UTC 8 December 2011 when the 710

cyclone center had crossed to northeast Scotland and the banding to the south was most

prominent. (b) Maximum 1-minute gusts at surface stations over central Scotland during 8 712

December 2011, filtered using a 10-minute median. Gust strength (ms–1) is colored by the time of

occurrence according to scale in upper left corner of the panel. 714

36

716

Figure 5. Path of the FAAM aircraft on 8 December 2011 with the track colored according to

altitude. Black dots indicate dropsonde launches. The flight took off from Exeter at 1048 UTC, 718

landed for refuelling in Teesside at 1607 UTC, took off again at 1729 UTC, and returned to Exeter

at 2110 UTC. The aircraft was at low levels within the strongest winds at around 1500 UTC and 720

again at 1900 UTC.

37

722

Figure 6. (Left panel) Cross sections derived from the ten dropsondes released along the first flight

leg from 1130 to 1234 UTC of: left panel, relative humidity with respect to ice (colors) and 724

potential temperature (white contours); right panel, wind speed (ms–1, colors ) and wind direction

(barbs); the wind barbs use the usual convention for wind strength in knots. Numbers in green 726

denote the order in which sondes were dropped and are placed at the corresponding latitude. In-

situ measurements from the aircraft, flying at a constant pressure of 390 hPa, are shown in the 728

strip at the top.

730

38

732

Figure 7. Precipitation-radar image for 1515 UTC showing the bands intercepted by the aircraft

(red track) between 1505 and 1517 UTC (rain rate mm hr–1, colored according to scale at bottom). 734

White lines: bands at 1500 UTC corresponding to the labels in Fig. 8a (B3 from shape of clouds

where precipitation is absent). Yellow lines: positions of the bands on the 1515 UTC image, 736

showing the southeastward progression of the bands. The southern bands S1 and S2 were not

intercepted by the aircraft. White dots on Tiree and Islay denote positions of automatic weather 738

stations.

740

39

742 Figure 8. Measurements from the FAAM aircraft as it flew northward from Islay to Tiree through

40

the strongest low-level winds. (a) Relative humidity with respect to ice along the flight track, 744

computed using WVSS-II data (shading) and rain rate (red line) derived from the radar network (interpolated to the flight track). (b) Wind speed (m s–1, black; the winds highlighted in yellow 746

correspond to regions where the droplet number concentration exceeds 20 cm-3 or the ice number concentration exceeds 5 l-1), radar altitude (x0.1km, grey dashed), temperature (°C, red) 748

and potential temperature (°C, blue). (c) Droplet number concentration (cm–3), as measured by the Cloud Droplet Probe. (d) Ice particle number concentration (l–1) as measured by the CIP-100 750

probe. (e) Diabatic heating and cooling rates associated with ice deposition and sublimation of ice crystals. Red line: mean value calculated over 8 s ( ~ 1 km) intervals using measured particle sizes 752

and shapes; grey shading: 1 Hz values, indicating the variability; blue line: as red line assuming spherical particles and exponential size distribution. Wind speed is shown again in black and 754

yellow, with an expanded scale

756

758

760

41

762

42

Figure 9. (a) Images of ice crystals captured by the Cloud Particle Imager (CPI) within cloud band 764

B1 (from 1510 – 1511 UTC), showing a mixture of columns and complex plate aggregates; (b) as (a)

in the middle of cloud band B2, revealing the dominance of columnar crystals, with evidence of 766

riming and aggregation; (c) comparison of particle size distributions (number of particles m-3 per

size bin, normalised to unit width) from the CIP-15 and CIP-100 probes, averaged over 1504–1516 768

UTC. The CIP-15 has a bin width of 15 μm and a maximum detectable size of 930 μm. The CIP-100

has a bin width of 100 μm and can detect precipitation-sized particles up to 6.2 mm. 770

43

772

774

Figure 10. A time–distance plot of radar-derived precipitation rate (mm hr–1, colored according to

scale at bottom) interpolated from the Met Office radar composite to a line connecting 776

observation sites at Tiree and Islay. Distance increases along the section from north-northwest to

south-southeast. Labels T and I identify the passage of rainband B1 over Tiree and Islay, 778

respectively, and point A indicates the crossing of this section by the aircraft. The time series of

wind gusts (black lines) measured at both AWS sites is overlain at the corresponding distance 780

along the section. A 90-minute running median has been removed from the winds to emphasise

the bands. The white curves indicate the progression of rainbands along the section (Fig. 8). 782

784

44

786 Figure 11. The 850-hPa wind speed (m s–1, colored according to scale at bottom) from the first four

members of the MOGREPS-UK trial forecast at 1600 UTC 8 December 2011, 7 h into the forecast. 788

Dashed lines indicate the axes of the wind maxima.

790

45

Figure 12. The rain rate (mm h–1, colored according to scale at bottom) from the first four 792

members of the MOGREPS-UK trial forecast at 1600 UTC 8 December 2011, 7 h into the forecast.

Red lines indicate the axes of the wind maxima identified in Fig. 11. 794

796

46

798

Figure 13. Fractions skill score (FSS) measuring the degree of fit between the rain rate pattern in each forecast and the radar data versus horizontal scale, averaged over forecast lead times 4–24 800

h. The yellow shading indicates the full range of results from the 12-member ensemble and the members 0, 4 and 5 are also indicated. The best ensemble member has a similar pattern to the 802

radar data (FSS > 0.5) for scales greater than 25 km. FSS is calculated over Scotland (54.6–59.3°N, 8.1°W–0.4°E). 804

806

808